library(magrittr)
library(tidyverse)
library(Seurat)
library(readxl)
library(cowplot)
library(colorblindr)
library(viridis)
library(progeny)

theme_cowplot2 <- function(...) {
  theme_cowplot(font_size = 12, ...) %+replace%
    theme(strip.background = element_blank(),
          plot.background = element_blank())
}
theme_set(theme_cowplot2())
coi <- params$cell_type
cell_sort <- params$cell_sort
cell_type_major <- params$cell_type_major
louvain_resolution <- params$louvain_resolution

1 Cluster markers

1.1 Major Ovarian.cancer.cell markers for cell assign

### load all data ---------------------------------

helper_f <- function(x) ifelse(is.na(x), "", x)
markers_v6 <- yaml::read_yaml("/home/uhlitzf/spectrum_tme/_data/small/signatures/hgsc_v6_major.yaml")

helper_f2 <- function(x) select(unnest(enframe(x, "subtype", "gene"), cols = gene), gene, subtype)
markers_v6_super <- lapply(yaml::read_yaml("/home/uhlitzf/spectrum_tme/_data/small/signatures/hgsc_v6_super.yaml"), helper_f2)

clrs <- yaml::read_yaml("/home/uhlitzf/spectrum_tme/_data/small/signatures/hgsc_v6_colors.yaml") %>% 
  lapply(function(x) map_depth(x, vec_depth(x)-2, unlist))

names(clrs$patient_id) <- str_remove_all(names(clrs$patient_id), "SPECTRUM-OV-")

meta_tbl <- read_excel("_data/small/MSK SPECTRUM - Single cell RNA-seq_v6.xlsx", sheet = 2) %>% 
  mutate(patient_id = str_remove_all(patient_id, "SPECTRUM-OV-")) %>% 
  filter(therapy == "pre-Rx")

signature_tbl <- read_tsv("_data/small/mutational_signatures_summary.tsv") %>% 
  mutate(patient_id = str_remove_all(patient_id, "SPECTRUM-OV-")) %>% 
  select(patient_id, consensus_signature) %>% 
  bind_rows(tibble(patient_id = unique(sort(meta_tbl$patient_id[!(meta_tbl$patient_id %in% .$patient_id)])), consensus_signature = "NA")) %>% 
  mutate(consensus_signature = ordered(consensus_signature, levels = names(clrs$consensus_signature))) %>%
  arrange(consensus_signature)

seu_obj <- read_rds("/work/shah/uhlitzf/data/SPECTRUM/freeze/v5/Ovarian.cancer.cell_highqc.rds")

myfeatures <- c("UMAP_1", "UMAP_2", "umapharmony_1", "umapharmony_2", "sample", "RNA_snn_res.0.1", "RNA_snn_res.0.2", "RNA_snn_res.0.3", "doublet", "nCount_RNA", "nFeature_RNA", "percent.mt", "doublet_score")

plot_data <- as_tibble(FetchData(seu_obj, myfeatures)) %>% 
  left_join(select(meta_tbl, sample = isabl_id, patient_id, tumor_supersite, tumor_subsite, sort_parameters, therapy), 
            by = "sample") %>% 
  mutate(patient_id = str_remove_all(patient_id, "SPECTRUM-OV-"),
         RNA_snn_res.0.1 = as.character(RNA_snn_res.0.1),
         RNA_snn_res.0.2 = as.character(RNA_snn_res.0.2),
         RNA_snn_res.0.3 = as.character(RNA_snn_res.0.3),
         tumor_supersite = ordered(tumor_supersite, levels = names(clrs$tumor_supersite))) %>% 
  mutate(cell_id = colnames(seu_obj)) %>% 
  left_join(signature_tbl, by = "patient_id")

patient_id <- sort(unique(plot_data$patient_id))

WFDC2, CD24, CLDN3, KRT7, KRT8, KRT17, KRT18, KRT19, EPCAM, WT1, CLDN4, MSLN, FOLR1, MUC1

1.2 Subtype currated markers

markers_v6_super[[coi]] %>% 
  group_by(subtype) %>% 
  mutate(rank = row_number(gene)) %>% 
  spread(subtype, gene) %>% 
  mutate_all(.funs = helper_f) %>% 
  formattable::formattable()

rank	Ciliated.cell.1	Ciliated.cell.2	Cycling.cancer.cell.1	Cycling.cancer.cell.2	doublet.Other	Endometrial.cell	Secretory.cell.1	Secretory.cell.2	Secretory.cell.3	Secretory.cell.4	Secretory.cell.5	Secretory.cell.6
1	C20orf85	KRT14	CENPF	ATP5MC1	AGR2	HP	KRT7	AGR2	FBXO32	AKAP12	CLDN1	CXCL10
2	CAPS	KRT16	HMGB2	PHB	CRISP3	OBP2A	PLCG2	CRISP3	GAS6	DCDC2	CLDN16	IFIT1
3	CAPSL	KRT5	MKI67	SNF8	IGFBP7	PAEP	SCGB2A1	OVGP1	MTSS1	MMP7	EREG	IFIT2
4	PIFO	MMP7	PTTG1	SNRPD1	ITM2B	PTGDS	SLPI	SCGB1A1	SAMHD1	TSPAN1	NDUFB5	IFIT3
5	TPPP3	SLURP1	TOP2A	TMEM106C	SPARCL1	SERINC2	SST	SCGB3A1	TFF3	TSPAN18		ISG15
6					TFF3			TFF3				MX1

1.3 Subtype cluster markers

marker_tbl <- read_tsv("/work/shah/uhlitzf/data/SPECTRUM/freeze/v5/Ovarian.cancer.cell_highqc_markers_02.tsv")

## Hypergeometric test --------------------------------------

test_set <- marker_tbl %>% 
  group_by(cluster) %>% 
  filter(!str_detect(gene, "^RPS|^RPL")) %>% 
  slice(1:50) %>% 
  mutate(k = length(cluster)) %>% 
  ungroup %>%
  select(cluster, gene, k) %>% 
  mutate(join_helper = 1) %>% 
  group_by(cluster, join_helper, k) %>%
  nest(test_set = gene)

markers_doub_tbl <- markers_v6 %>% 
  enframe("subtype", "gene") %>% 
  filter(!(subtype %in% unique(c(coi, cell_type_major)))) %>% 
  unnest(gene) %>% 
  group_by(gene) %>% 
  filter(length(gene) == 1) %>% 
  mutate(subtype = paste0("doublet.", subtype)) %>% 
  bind_rows(tibble(subtype = "Mito.high", gene = grep("^MT-", rownames(seu_obj), value = T)))

ref_set <- markers_v6_super[[coi]] %>% 
  bind_rows(markers_doub_tbl) %>% 
  group_by(subtype) %>% 
  mutate(m = length(gene),
         n = length(rownames(seu_obj))-m,
         join_helper = 1) %>% 
  group_by(subtype, m, n, join_helper) %>%
  nest(ref_set = gene)

hyper_tbl <- test_set %>% 
  left_join(ref_set, by = "join_helper") %>% 
  group_by(cluster, subtype, m, n, k) %>%
  do(q = length(intersect(unlist(.$ref_set), unlist(.$test_set)))) %>%
  mutate(pval = 1-phyper(q = q, m = m, n = n, k = k)) %>%
  ungroup %>%
  mutate(qval = p.adjust(pval, "BH"),
         sig = qval < 0.01)

# hyper_tbl %>% 
#   group_by(subtype) %>% 
#   filter(any(qval < 0.01)) %>%
#   ggplot(aes(subtype, -log10(qval), fill = sig)) +
#   geom_bar(stat = "identity") +
#   facet_wrap(~cluster) +
#   coord_flip()
  
low_rank <- str_detect(unique(hyper_tbl$subtype), "doublet")
subtype_lvl <- c(sort(unique(hyper_tbl$subtype)[!low_rank]), sort(unique(hyper_tbl$subtype)[low_rank]))
  
cluster_label_tbl <- hyper_tbl %>% 
  mutate(subtype = ordered(subtype, levels = subtype_lvl)) %>% 
  arrange(qval, subtype) %>%
  group_by(cluster) %>% 
  slice(1) %>% 
  mutate(subtype = ifelse(sig, as.character(subtype), paste0("unknown_", cluster))) %>% 
  select(cluster, cluster_label = subtype) %>% 
  ungroup %>% 
  mutate(cluster_label = make.unique(cluster_label, sep = "_"))

seu_obj$cluster_label <- unname(deframe(cluster_label_tbl)[as.character(unlist(seu_obj[[paste0("RNA_snn_res.", louvain_resolution)]]))])
plot_data$cluster_label <- seu_obj$cluster_label

marker_sheet <- marker_tbl %>% 
  left_join(cluster_label_tbl, by = "cluster") %>% 
  group_by(cluster_label) %>% 
  filter(!str_detect(gene, "^RPS|^RPL")) %>% 
  slice(1:50) %>% 
  mutate(rank = row_number(-avg_logFC)) %>% 
  select(cluster_label, gene, rank) %>% 
  spread(cluster_label, gene) %>% 
  mutate_all(.funs = helper_f)

formattable::formattable(marker_sheet)

rank	Ciliated.cell.1	Ciliated.cell.1_1	Cycling.cancer.cell.1	Cycling.cancer.cell.1_1	Cycling.cancer.cell.2	doublet.Fibroblast	doublet.Monocyte	doublet.Monocyte_1	doublet.Other	doublet.Plasma.cell	Endometrial.cell	Secretory.cell.1	Secretory.cell.1_1	Secretory.cell.2	Secretory.cell.2_1	Secretory.cell.5	Secretory.cell.6	unknown_13
1	C20orf85	CFAP126	HIST1H4C	CENPF	PHB	TAGLN	IGKC	APOE	CRISP3	IGLC2	PAEP	SCGB2A1	S100A9	OVGP1	SPINK1	CLDN16	CXCL10	NTS
2	CAPS	DNALI1	TUBA1B	PTTG1	SNF8	TPM2	IGLC2	NDUFA4L2	IGFBP7	IGKC	HP	DAPL1	SLPI	CRISP3	CXCL1	EREG	ISG15	CALB1
3	TPPP3	PIFO	PCLAF	UBE2S	ATP5MC1	LGALS1	IGLC3	DMKN	ITM2B	CRIP1	PTGDS	SCGB1D2	TACSTD2	PLCG2	SFTA2	PCLAF	IFIT3	MAOB
4	RSPH1	FAM183A	ATAD2	HMGB2	MEG3	ACTA2	IGHG1	LINC00665	SCGB1D2	HIST1H4C	SCGB2A1	SNHG19	LCN2	CCND1	CXCL8	HIST1H1A	IFIT1	SOX18
5	C9orf24	CASC1	MKI67	TPX2	TMBIM4	CALD1	SPP1	COX6B1	EIF4A2	IGLC3	OBP2A	SNHG8	S100A10	PTMS	CST6	RAC3	MX1	CDH2
6	C5orf49	RSPH1	H2AFZ	ASPM	RARRES2	MT1H	SRGN	GPI	EPB41L4A-AS1	TOP2A	PCP4	KCNQ1OT1	IGFBP3	SNHG9	CXCL3	CTHRC1	IFIT2	CHST1
7	PIFO	LRRIQ1	PCNA	MKI67	UBE2Z	NMU	C1QB	C19orf53	CIRBP	ATP1B3	S100A4	GABPB1-AS1	GPRC5A	DSTN	CAMK2N1	TMEM100	PARP14	MFSD3
8	TMEM190	C20orf85	TK1	CDC20	PDK2	RAB32	CCL4	PEPD	ASRGL1	DEFB1	SERINC2		KRT17	PEG10	SPP1	MMP23B	IFI6	RSPO3
9	FAM183A	DNAAF1	TUBB	TOP2A	ITM2C	SELENOP	GNLY	PSENEN	PIK3R1	SNRPF	CRABP1		KRT7	YWHAE	UPK1B	RRM2	IFI44L	PIR
10	C1orf194	C5orf49	CKS1B	CKS2	SLC35B1	ANXA3	RGS1	PDCD5	EEF1A1	HMGN1	FOLR1		RARRES1	APOLD1	AGR2	SYT10	GBP4	AL589182.1
11	CETN2	C9orf24	TYMS	CCNB1	PCP4	MT1G	IGHG3	USP11	SNHG7	KLK5	SCGB1D1		KRT19	CDC20	CAV1	PHOX2A	STAT1	MHENCR
12	AGR3	AGR3	DUT	BIRC5	XPOT	CDKN2A	C1QA	TMEM147	NPM1	IGHG1	HLA-DQA1		VEGFA	EPB41L2	CXCL5	CNIH2	TNFSF10	AP000527.1
13	TXN	C1orf194	HMGB2	ARL6IP1	ZCRB1	CCT5	LYZ	IGF2.1	TPT1	NNMT	HLA-DRB5		SOX4	MYL12B	GNG11	C1QTNF12	GBP1	SIX1
14	ODF3B	ARMC3	TOP2A	LGALS1	SPOP	VIM	CCL3	FBXO2	BTF3	IGF1	SNCG		MMP7	PTPRA	ADM	TMEM178A	XAF1	PRRX2
15	CAPSL	CFAP45	UBE2C	UBE2C	ZNF652	YBX1	KRT6A	NPW	FAU	ATP5ME	HLA-DQB1		NDRG1	RBMX	AC092069.1	AC008771.1	SOD2	PREX2
16	CFAP126	EFCAB1	SMC4	JPT1	HOXB4	AKAP12	IGFL2-AS1	CLDN9	EHF	HMGB1	AC007906.2		S100A6	PABPC4	CXCL2	MTRNR2L1	IDO1	PYCR3
17	TUBA1A	TPPP3	DEK	CEP55	HOXB2	COTL1	TYROBP	CPVL	KCNMA1	PCSK1N	HSPA1A		FTL	METAP2	PTGS2	CENPM	OAS1	FHL1
18	MORN2	ZMYND10	MCM3	CCNB2	FXYD1	IGFBP2	FCER1G	MFAP4	CAPS	C4orf48	CHI3L1		CRYAB	VIM	TNFRSF11B	CENPH	IFI27	MPPED2
19	C11orf88	PPIL6	HELLS	PTMS	TWF1	FHL2	PTPRC	GSTM3	AR	RARRES1	CYP4B1		ANXA1	MYL6B	RHOBTB3	CENPK	TNFAIP2	F3
20	DNAAF1	ENKUR	STMN1	NMU	C2orf40	NDUFS6	FCGR3A	HMGA2	UGT2B17	ATP5PF	HLA-DRA		SLC2A1	RBM8A	VGLL3	PRTFDC1	RSAD2	ACSM3
21	TUBB4B	CDC20B	CDK1	CDKN3	KAT7	PCDH10	CCL4L2	FXYD1	GYPC	HIST1H1C	DEFB1		TFPI2	SNHG19	MALL	POMC	IFIH1	MMD
22	FOXJ1	SPATA17	NUSAP1	TUBA1B	AC008124.1	EEF1B2	CCL5	GFRA1	EEF2	S100A10	HLA-B		FTH1	WBP11	EFEMP1	GPT	PSMB9	NDP
23	ZMYND10	CAPSL	MCM7	NUSAP1	COMP	UCHL1	LAPTM5	ZFP14	EIF1	DEK	S100A1		CD55	PPT1	WNT2	MELTF-AS1	IL32	AP001528.2
24	LRRIQ1	TEKT2	CENPF	H2AFZ	LHX1	MYH9	C1QC	HIST1H2BC	ERP27	UBE2S	HLA-DPA1		C19orf33	BRD9	KRT13	SPC24	CXCL11	APOO
25	IGFBP7	CFAP44	CLSPN	KPNA2	SEMA3E	MYL12A	IL7R	COL9A3	TMEM98	MT1E	CITED4		S100A11	ST13	IL20RA	MYBL2	DDX58	GBP5
26	IK	TEKT1	MCM4	HMGB1	CRTAC1	SUB1	ITGB2	MXRA8	SORBS2	NUSAP1	NPDC1		TNFRSF12A	SNHG25	ANKRD1	OXTR	SAMD9	RHBDD3
27	EFCAB1	CFAP54	UBE2T	HMGN2	FBXO2	ANXA5	CD52	TIMP3	HOXB4	MKI67	PRSS33		EMP1	AMD1	LINC01127	AC025159.1	RNF213	SAMD10
28	DYDC2	CFAP43	MYBL2	ECT2	LEMD3	MYL9	IGHGP	BGN	PART1	CCL4	GMPR		PI3	SNRPB	CXCL6	AFAP1-AS1	IRF1	NRARP
29	EFHC1	SPEF1	SMC2	ANP32E	HOXB-AS1	CLDN7	CD14	IGF2BP3	FYB2	SMC1B	HLA-DRB1		ANXA2	NAP1L1	RAB42	ASF1B	WARS	LINC02303
30	DNALI1	FAM81B	HMGB1	CENPE	LGR5	S100A10	AIF1	CA9	FRMD4A	IGHG3	GADD45B		MAL2	SNRPB2	GABRP	PTPRS	CCL20	SLC28A2
31	AL357093.2	EFCAB10	RAD51AP1	STMN1	LMO3	CD151	FYB1	CPS1	SYNE1	RGS1	MESP1		S100A2	PA2G4	XAGE2	CALML5	B2M	SOX2
32	C9orf116	CFAP46	TPX2	HMMR	IGF2BP1	MT1M	CCL3L1	FBN2	ALDH1A1	PTMA	FAM107A		TAGLN	NSA2	C5orf46	LINC01436	CXCL8	ANKRD6
33	CEP126	ROPN1L	ZWINT	HMGB3	IL11RA	CD47	NKG7	STC2	PDE4D	C1QB	MFGE8		CAMK2N1	NMT1	ABLIM3	HOMER1	ISG20	MAN1B1-DT
34	SNTN	CFAP73	RRM2	TROAP	COL9A3	FLNA	ALOX5AP	PLOD1	HSPB6	SLC16A8	CYBA		LDHA	TMCO1	COL4A4	F2R	TAP1	KCNMB4
35	CCDC170	C7orf57	NASP	KIF20B	MDM1	IL32	SAMSN1	NRN1	HOXB6	SRGN	IRX3		AKAP12	WDR70	COL8A1	SEMA6A	LAP3	TBX2
36	PERP	SPAG8	CENPW	SMC4	TSPOAP1	MYL12B	CD2	PAGE2	KAZALD1	APOC1	MYL9		TGM2	IGFBP3	GDF7	PTGER1	OAS3	ABO
37	ROPN1L	WDR38	CENPX	CKS1B	CLDN9	PSMB6	IGHG4	HCST	SYT10	FAM111B	PRDX5		ITGB8	TBCA	CCL20	FZD8	HLA-B	BMPR1B-DT
38	CKB	WDR49	TMPO	NUCKS1	PRELP	ACTR3	GPR183	HOXA10-AS	TGFB2	AIF1	NUPR1		BNIP3	KRT81	AC245041.2	HACD1	IFI16	FES
39	TEKT2	ZBBX	E2F1	H2AFV	C12orf56	PDCD6	LSP1	FGFR3	SH3BP5	IGHG4	CLDN3		SLC40A1	TTYH1	BMP4	SALL4	LY6E	SLC8A1
40	C9orf135	DRC1	RANBP1	RAD21	AC090673.1	SRI	CD3D	CPXM2	MAP3K1	TYROBP	SLC26A2		SQSTM1	HMGB1	ALPP	CENPN	PLSCR1	AL713998.1
41	SPA17	WDR63	DNMT1	NCAPD2	FXYD6	NDUFS5	CD37	HOTAIRM1	AKR1C3	CCL5	HOMER3		CD24	ATP5ME	CDA	SOX15	HLA-A	ZNF252P-AS1
42	GDF15	CFAP157	MAD2L1	DLGAP5	FAM117A	COX4I1	IL1B	NXPH4	DPP4	TCHH	HLA-DMA		EZR	AGR3	MARCH10	CDC6	NFKBIA	AC064807.1
43	TEKT1	DYNLRB2	TMEM106C	AURKA	CCDC170	EEF1D	GMFG	KRBOX1	ACSM3	C1QA	MGLL		FAM3C	TMA7	GPR87	FANCD2	HLA-C	CPED1
44	PSENEN	C6orf118	HIST1H1D	CENPW	COL6A2	TPM3	MS4A7	LRIG3	SLITRK5	FCGR3A	HLA-DPB1		MUC4	SFRP1	DKK1	CCR10	IFI44	THAP7-AS1
45	WDR38	RSPH4A	ASPM	NEK2	CYYR1	UQCRH	CD7	SDC2	PREX2	DDIT4L	ETV4		ATP1A1	TPT1	SLC16A12	CACNG4	OAS2	UFSP1
46	LRRC23	DNAH12	FANCI	TUBB4B	CRYGD	CXCL12	MS4A6A	TMEM147-AS1	CMBL	FLG	ATP6V0B		CLDN1	ANO1	SULT1E1	MND1	BIRC3	EFHC2
47	ARMC3	TTLL9	BIRC5	PLK1	GPC6	PDCL2	LST1	AEBP1	CCND2	CD2	UCP2		P4HA1	MEG3	THBD	CASC10	HERC5	SYT12
48	MLF1	MAP3K19	ORC6	CENPA	FGF17	NPTX1	GZMB	FGF11	CCDC170	PTPRC	SPDEF		CST6	CCND2	MSX2	BHMT2	UBE2L6	TBX3
49	ERICH3	MORN5	DTYMK	MAD2L1	PLXNA4	MUC12	GZMA	KCNG1	IL18	IGHGP	JUND		SLC16A3	CD177	GATA3	AC144450.1	SAMD9L	TMEM135
50	CFAP45	STMND1	TUBB4B	TUBA1C	TBX2-AS1	HPD	CD53	TGFB1	SKAP1	TRBC1	EMP2		NEAT1	PEG3	CYP2W1	CGAS	CXCL2	AL161669.3

write_tsv(marker_sheet, paste0("/work/shah/uhlitzf/data/SPECTRUM/freeze/v6/", coi, "_marker_sheet.tsv"))

2 Clusters

2.1 sizes

enframe(sort(table(seu_obj$cluster_label))) %>% 
  mutate(name = ordered(name, levels = rev(name))) %>% 
  ggplot() +
  geom_bar(aes(name, value), stat = "identity") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
  labs(y = c("#cells"), x = "cluster")

2.2 UMAP

alpha_lvl <- ifelse(nrow(plot_data) < 20000, 0.2, 0.1)
pt_size <- ifelse(nrow(plot_data) < 20000, 0.2, 0.05)

common_layers_disc <- list(  
  geom_point(size = pt_size, alpha = alpha_lvl),
  NoAxes(),
  guides(color = guide_legend(override.aes = list(size = 2, alpha = 1))),
  labs(color = "")
)

common_layers_cont <- list(  
  geom_point(size = pt_size, alpha = alpha_lvl),
  NoAxes(),
  scale_color_gradientn(colors = viridis(9)),
  guides(color = guide_colorbar())
)

ggplot(plot_data, aes(umapharmony_1, umapharmony_2, color = cluster_label)) + 
  common_layers_disc +
  #facet_wrap(~therapy) +
  ggtitle("Sub cluster")

3 Filtering out doublet and mito high clusters

3.1 UMAP

my_subtypes <- names(clrs$cluster_label[[coi]])

my_subtypes <- c(my_subtypes, unlist(lapply(paste0("_", 1:3), function(x) paste0(my_subtypes, x)))) %>% .[!str_detect(., "doublet")]

cells_to_keep <- colnames(seu_obj)[seu_obj$cluster_label %in% my_subtypes]
# seu_obj_sub <- subset(seu_obj, cells = cells_to_keep)
# seu_obj_sub <- RunUMAP(seu_obj_sub, dims = 1:20, reduction = "harmony", reduction.name = "umapharmony", reduction.key = "umapharmony_")
# seu_obj_sub$cluster_label <- seu_obj$cluster_label[colnames(seu_obj) %in% colnames(seu_obj_sub)]
# write_rds(seu_obj_sub, paste0("/work/shah/uhlitzf/data/SPECTRUM/freeze/v6/", coi, "_processed_filtered.rds"))
seu_obj_sub <- read_rds(paste0("/work/shah/uhlitzf/data/SPECTRUM/freeze/v6/", coi, "_processed_filtered.rds"))

plot_data_sub <- as_tibble(FetchData(seu_obj_sub, c(myfeatures, "cluster_label"))) %>% 
  left_join(select(meta_tbl, sample = isabl_id, patient_id, tumor_supersite, tumor_subsite, sort_parameters, therapy), 
            by = "sample") %>% 
  mutate(patient_id = str_remove_all(patient_id, "SPECTRUM-OV-"),
         tumor_supersite = ordered(tumor_supersite, levels = names(clrs$tumor_supersite))) %>% 
  mutate(cell_id = colnames(seu_obj_sub),
         cluster_label = ordered(cluster_label, levels = my_subtypes),
         ) %>% 
  left_join(signature_tbl, by = "patient_id")
  
if (cell_sort == "CD45+") {
plot_data_sub <- filter(plot_data_sub, sort_parameters != "singlet, live, CD45-", !is.na(tumor_supersite))
}

if (cell_sort == "CD45-") {
plot_data_sub <- filter(plot_data_sub, sort_parameters != "singlet, live, CD45+", !is.na(tumor_supersite))
}

ggplot(plot_data_sub, aes(umapharmony_1, umapharmony_2, color = cluster_label)) + 
  common_layers_disc +
  # scale_color_manual(values = clrs$cluster_label[[coi]])
  #facet_wrap(~cluster_label) +
  ggtitle("Sub cluster")

ggplot(plot_data_sub, aes(umapharmony_1, umapharmony_2, color = patient_id)) + 
  common_layers_disc +
  ggtitle("Patient") +
  # facet_wrap(~therapy) +
  scale_color_manual(values = clrs$patient_id)

ggplot(plot_data_sub, aes(umapharmony_1, umapharmony_2, color = tumor_supersite)) + 
  # geom_point(aes(umapharmony_1, umapharmony_2), 
  #            color = "grey90", size = 0.01, 
  #            data = select(plot_data_sub, -tumor_supersite)) +
  common_layers_disc +
  ggtitle("Site") +
  # facet_wrap(~therapy) +
  scale_color_manual(values = clrs$tumor_supersite)

write_tsv(select(plot_data_sub, cell_id, everything(), -UMAP_1, -UMAP_2, -contains("RNA_")), paste0("/work/shah/uhlitzf/data/SPECTRUM/freeze/v6/", coi, "_embedding.tsv"))

3.2 add module scores and pathway scores

signature_modules <- read_excel("_data/small/signatures/SPECTRUM v5 sub cluster markers.xlsx", sheet = 2, skip = 1, range = "M2:P100") %>% 
  gather(module, gene) %>% 
  na.omit() %>% 
  group_by(module) %>% 
  do(gene = c(.$gene)) %>% 
  {setNames(.$gene, .$module)}

signature_modules$ISG.module <- c("CCL5", "CXCL10", "IFNA1", "IFNB1", "ISG15", "IFI27L2", "SAMD9L")

## compute expression module scores
for (i in 1:length(signature_modules)) {
  seu_obj_sub <- AddModuleScore(seu_obj_sub, features = signature_modules[i], name = names(signature_modules)[i])
  seu_obj_sub[[names(signature_modules)[i]]] <- seu_obj_sub[[paste0(names(signature_modules)[i], "1")]]
  seu_obj_sub[[paste0(names(signature_modules)[i], "1")]] <- NULL
  print(paste(names(signature_modules)[i], "DONE"))
}

## [1] "CD8.Cytotoxic DONE"
## [1] "CD8.Dysfunctional DONE"
## [1] "CD8.Naive DONE"
## [1] "CD8.Predysfunctional DONE"
## [1] "ISG.module DONE"

## compute progeny scores
progeny_list <- seu_obj_sub@assays$RNA@data[VariableFeatures(seu_obj_sub),] %>% 
  as.matrix %>% 
  progeny %>% 
  as.data.frame %>% 
  as.list

names(progeny_list) <- make.names(paste0(names(progeny_list), ".pathway"))

for (i in 1:length(progeny_list)) {
  seu_obj_sub <- AddMetaData(seu_obj_sub, metadata = progeny_list[[i]], 
                             col.name = names(progeny_list)[i])
}

write_rds(seu_obj_sub, paste0("/work/shah/uhlitzf/data/SPECTRUM/freeze/v6/", coi, "_processed_filtered.rds"))

3.3 marker heatmap

marker_top_tbl <- marker_sheet[,-1] %>% 
  slice(1:10) %>% 
  as.list %>% 
  .[!str_detect(names(.), "doublet")] %>% 
  enframe("cluster_label_x", "gene") %>% 
  unnest(gene)

plot_data_markers <- as_tibble(FetchData(seu_obj_sub, c("cluster_label", myfeatures, unique(marker_top_tbl$gene)))) %>% 
  gather(gene, value, -c(1:(length(myfeatures)+1))) %>% 
  left_join(select(meta_tbl, sample = isabl_id, patient_id, tumor_supersite, tumor_subsite, sort_parameters, therapy), 
            by = "sample") %>% 
  mutate(patient_id = str_remove_all(patient_id, "SPECTRUM-OV-"),
         tumor_supersite = ordered(tumor_supersite, levels = names(clrs$tumor_supersite))) %>% 
  mutate(cluster_label = ordered(cluster_label, levels = my_subtypes)) %>% 
  group_by(cluster_label, gene) %>% 
  summarise(value = mean(value, na.rm = T)) %>% 
  group_by(gene) %>% 
  mutate(value = scales::rescale(value)) %>% 
  left_join(marker_top_tbl, by = "gene") %>% 
  mutate(cluster_label_x = ordered(cluster_label_x, levels = rev(names(clrs$cluster_label[[coi]]))))

ggplot(plot_data_markers) +
  geom_tile(aes(gene, cluster_label, fill = value)) +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
  facet_grid(~cluster_label_x, scales = "free", space = "free") +
  scale_fill_gradientn(colors = viridis(9)) +
  labs(fill = "Scaled\nexpression") +
  theme(aspect.ratio = 1,
        axis.line = element_blank(),
        axis.ticks = element_blank(),
        axis.title = element_blank())

# ggsave(paste0("_fig/002_marker_heatmap_", coi, ".pdf"), width = nrow(marker_top_tbl)/6, height = 5)

3.4 composition

3.4.1 per site

comp_site_tbl <- plot_data_sub %>%
  filter(!is.na(tumor_supersite)) %>% 
  group_by(cluster_label, tumor_supersite) %>%
  tally %>%
  group_by(tumor_supersite) %>%
  mutate(nrel = n/sum(n)*100) %>%
  ungroup

pnrel_site <- ggplot(comp_site_tbl) +
  geom_bar(aes(tumor_supersite, nrel, fill = cluster_label),
           stat = "identity", position = position_stack()) +
  scale_y_continuous(expand = c(0, 0)) +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
  # scale_fill_manual(values = clrs$cluster_label[[coi]]) +
  labs(fill = "Cluster", y = "Fraction [%]", x = "")

pnabs_site <- ggplot(comp_site_tbl) +
  geom_bar(aes(tumor_supersite, n, fill = cluster_label),
           stat = "identity", position = position_stack()) +
  scale_y_continuous(expand = c(0, 0)) +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
  # scale_fill_manual(values = clrs$cluster_label[[coi]]) +
  labs(fill = "Cluster", y = "# cells", x = "") 

plot_grid(pnabs_site, pnrel_site, ncol = 2, align = "h")

# ggsave(paste0("_fig/02_deep_dive_", coi, "_comp_site.pdf"), width = 8, height = 4)

3.4.2 per sample

comp_tbl_sample_sort <- plot_data_sub %>% 
  group_by(tumor_subsite, tumor_supersite, patient_id, consensus_signature, therapy, cluster_label) %>% 
  tally %>% 
  group_by(tumor_subsite, tumor_supersite, patient_id, consensus_signature, therapy) %>% 
  mutate(nrel = n/sum(n)*100,
         nsum = sum(n),
         log10n = log10(n),
         label_supersite = "Site",
         label_mutsig = "Signature",
         label_therapy = "Rx") %>% 
  ungroup %>% 
  arrange(desc(therapy), tumor_supersite) %>% 
  mutate(tumor_subsite_rx = paste0(tumor_subsite, "_", therapy)) %>% 
  mutate(tumor_subsite = ordered(tumor_subsite, levels = unique(tumor_subsite)),
         tumor_subsite_rx = ordered(tumor_subsite_rx, levels = unique(tumor_subsite_rx))) %>% 
  arrange(patient_id) %>% 
  mutate(label_patient_id = ifelse(as.logical(as.numeric(fct_inorder(as.character(patient_id)))%%2), "Patient1", "Patient2"))

sample_id_x_tbl <- plot_data_sub %>% 
  mutate(sort_short_x = cell_sort) %>% 
  distinct(patient_id, sort_short_x, tumor_subsite, therapy, sample) %>% 
  unite(sample_id_x, patient_id, sort_short_x, tumor_subsite, therapy) %>% 
  arrange(sample_id_x)

comp_tbl_sample_sort %>% 
  mutate(sort_short_x = cell_sort) %>% 
  unite(sample_id_x, patient_id, sort_short_x, tumor_subsite_rx) %>% 
  select(sample_id_x, cluster_label, n, nrel, nsum) %>% 
  left_join(sample_id_x_tbl, by = "sample_id_x") %>% 
  write_tsv(paste0("/work/shah/uhlitzf/data/SPECTRUM/freeze/v6/", coi, "_subtype_compositions.tsv"))

ybreaks <- c(500, 1000, 2000, 4000, 6000, 8000, 10000, 15000, 20000)

max_cells_per_sample <- max(comp_tbl_sample_sort$nsum)
ymaxn <- ybreaks[max_cells_per_sample < ybreaks][1]

comp_plot_wrapper <- function(y = "nrel", switch = NULL) {
  if (y == "nrel") ylab <- paste0("Fraction\nof cells [%]")
  if (y == "n") ylab <- paste0("Number\nof cells")
  p <- ggplot(comp_tbl_sample_sort, 
              aes_string("tumor_subsite_rx", y, fill = "cluster_label")) + 
    facet_grid(~patient_id, space = "free", scales = "free", switch = switch) +
    coord_cartesian(clip = "off") + 
    # scale_fill_manual(values = clrs$cluster_label[[coi]]) + 
    theme(axis.text.x = element_blank(),
          axis.title.y = element_text(angle = 0, vjust = 0.5, hjust = 0.5, 
                                      margin = margin(0, -0.4, 0, 0, unit = "npc")),
          axis.ticks.x = element_blank(),
          axis.title.x = element_blank(),
          axis.line.x = element_blank(),
          strip.text.y = element_blank(),
          strip.text.x = element_blank(),
          strip.background.y = element_blank(),
          strip.background.x = element_blank(),
          plot.margin = margin(0, 0, 0, 0, "npc")) + 
    labs(x = "",
         y = ylab) +
    guides(fill = FALSE)
  if (y == "nrel") p <- p + 
    geom_bar(stat = "identity") +
    scale_y_continuous(expand = c(0, 0), 
                       breaks = c(0, 50, 100), 
                       labels = c("0", "50", "100"))
  if (y == "n") p <- p + 
    geom_bar(stat = "identity", position = position_stack()) +
    scale_y_continuous(expand = c(0, 0), 
                       breaks = c(0, ymaxn/2, ymaxn),
                       limits = c(0, ymaxn),
                       labels = c("", ymaxn/2, ymaxn)) +
    expand_limits(y = c(0, ymaxn)) +
    theme(panel.grid.major.y = element_line(linetype = 1, color = "grey90", size = 0.5))
  return(p)
} 

common_label_layers <- list(
  geom_tile(color = "white", size = 0.15),
  theme(axis.text.x = element_blank(),
        axis.ticks = element_blank(),
        axis.title.x = element_blank(),
        axis.line.x = element_blank(),
        strip.text = element_blank(),
        strip.background = element_blank(),
        plot.margin = margin(0, 0, 0, 0, "npc")),
  scale_y_discrete(expand = c(0, 0)),
  labs(y = ""),
  guides(fill = FALSE),
  facet_grid(~patient_id, 
             space = "free", scales = "free")
)

comp_label_site <- ggplot(distinct(comp_tbl_sample_sort, tumor_subsite_rx, therapy, tumor_supersite, label_supersite, patient_id), 
       aes(tumor_subsite_rx, label_supersite, 
           fill = tumor_supersite)) + 
  scale_fill_manual(values = clrs$tumor_supersite) +
  common_label_layers

comp_label_rx <- ggplot(distinct(comp_tbl_sample_sort, tumor_subsite_rx, therapy, tumor_supersite, label_therapy, consensus_signature, patient_id), 
       aes(tumor_subsite_rx, label_therapy, 
           fill = therapy)) + 
  scale_fill_manual(values = c(`post-Rx` = "gold3", `pre-Rx` = "steelblue")) +
  common_label_layers

comp_label_mutsig <- ggplot(distinct(comp_tbl_sample_sort, tumor_subsite_rx, therapy, tumor_supersite, label_mutsig, consensus_signature, patient_id), 
       aes(tumor_subsite_rx, label_mutsig, 
           fill = consensus_signature)) + 
  scale_fill_manual(values = clrs$consensus_signature) +
  common_label_layers

patient_label_tbl <- distinct(comp_tbl_sample_sort, patient_id, .keep_all = T)

comp_label_patient_id <- ggplot(patient_label_tbl, aes(tumor_subsite_rx, label_patient_id)) + 
  scale_fill_manual(values = clrs$consensus_signature) +
  geom_text(aes(tumor_subsite_rx, label_patient_id, label = patient_id)) +
  facet_grid(~patient_id, 
             space = "free", scales = "free") +
  coord_cartesian(clip = "off") + 
  theme_void() +
  theme(strip.text = element_blank(),
        strip.background = element_blank(),
        plot.margin = margin(0, 0, 0, 0, "npc"))

hist_plot_wrapper <- function(x = "nrel") {
  if (x == "nrel") {
    xlab <- paste0("Fraction of cells [%]")
    bw <- 5
  }
  if (x == "log10n") {
    xlab <- paste0("Number of cells")
    bw <- 0.2
  }
  p <- ggplot(comp_tbl_sample_sort) +
    ggridges::geom_density_ridges(
      aes_string(x, "cluster_label", fill = "cluster_label"), color = "black",
      stat = "binline", binwidth = bw, scale = 3) +
    facet_grid(label_supersite~., 
               space = "free", scales = "free") +
    # scale_fill_manual(values = clrs$cluster_label[[coi]]) + 
    theme(axis.text.y = element_blank(),
          axis.ticks.y = element_blank(),
          axis.title.y = element_blank(),
          axis.line.y = element_blank(),
          strip.text = element_blank(),
          strip.background = element_blank(),
          plot.margin = margin(0, 0, 0, 0, "npc")) +
    scale_x_continuous(expand = c(0, 0)) +
    scale_y_discrete(expand = c(0, 0)) +
    guides(fill = FALSE) +
    labs(x = xlab)
  if (x == "log10n") p <- p + expand_limits(x = c(0, 3)) + 
    scale_x_continuous(expand = c(0, 0), 
                       labels = function(x) parse(text = paste("10^", x)))
  return(p)
}

pcomp1 <- comp_plot_wrapper("n")
pcomp2 <- comp_plot_wrapper("nrel")

pcomp_grid <- plot_grid(comp_label_patient_id, 
                        pcomp1, pcomp2, 
                        comp_label_site, comp_label_rx, comp_label_mutsig,
                        ncol = 1, align = "v", 
                        rel_heights = c(0.15, 0.33, 0.33, 0.06, 0.06, 0.06))

phist1 <- hist_plot_wrapper("log10n")

pcomp_hist_grid <- ggdraw() +
  draw_plot(pcomp_grid, x = 0.01, y = 0, width = 0.85, height = 1) +
  draw_plot(phist1, x = 0.87, y = 0.05, width = 0.12, height = 0.8)

pcomp_hist_grid

# ggsave(paste0("_fig/02_composition_v6_",coi,".pdf"), pcomp_hist_grid, width = 10, height = 2)

3.4.3 site specific cluster enrichment

comp_tbl_z <- comp_tbl_sample_sort %>% 
  filter(therapy == "pre-Rx",
         !(tumor_supersite %in% c("Ascites", "Other"))) %>% 
  group_by(patient_id, cluster_label) %>% 
  arrange(patient_id, cluster_label, nrel) %>% 
  mutate(rank = row_number(nrel),
         z_rank = scales::rescale(rank)) %>% 
  mutate(mean_nrel = mean(nrel, na.rm = T),
         sd_nrel = sd(nrel, na.rm = T),
         z_nrel = (nrel - mean_nrel) / sd_nrel) %>% 
  ungroup()

ggplot(comp_tbl_z) +
  geom_boxplot(aes(tumor_supersite, z_nrel, color = tumor_supersite), 
               outlier.shape = NA) +
  geom_point(aes(tumor_supersite, z_nrel, color = tumor_supersite), position = position_jitter(), size = 0.1) +
  facet_grid(~cluster_label, scales = "free_x", space = "free_x") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5),
        aspect.ratio = 5) +
  scale_color_manual(values = clrs$tumor_supersite)

ggplot(comp_tbl_z) +
  geom_boxplot(aes(tumor_supersite, z_rank, color = tumor_supersite), 
               outlier.shape = NA) +
  geom_point(aes(tumor_supersite, z_rank, color = tumor_supersite), position = position_jitter(), size = 0.1) +
  facet_grid(~cluster_label, scales = "free_x", space = "free_x") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5),
        aspect.ratio = 5) +
  scale_color_manual(values = clrs$tumor_supersite)

4 session info

devtools::session_info()

## ─ Session info ───────────────────────────────────────────────────────────────
##  setting  value                       
##  version  R version 3.6.2 (2019-12-12)
##  os       Debian GNU/Linux 10 (buster)
##  system   x86_64, linux-gnu           
##  ui       X11                         
##  language (EN)                        
##  collate  en_US.UTF-8                 
##  ctype    en_US.UTF-8                 
##  tz       Etc/UTC                     
##  date     2020-12-05                  
## 
## ─ Packages ───────────────────────────────────────────────────────────────────
##  package      * version    date       lib
##  ape            5.3        2019-03-17 [2]
##  assertthat     0.2.1      2019-03-21 [2]
##  backports      1.1.10     2020-09-15 [1]
##  bibtex         0.4.2.2    2020-01-02 [2]
##  Biobase        2.46.0     2019-10-29 [2]
##  BiocGenerics   0.32.0     2019-10-29 [2]
##  bitops         1.0-6      2013-08-17 [2]
##  broom          0.7.2      2020-10-20 [1]
##  callr          3.4.2      2020-02-12 [1]
##  caTools        1.17.1.4   2020-01-13 [2]
##  cellranger     1.1.0      2016-07-27 [2]
##  cli            2.0.2      2020-02-28 [1]
##  cluster        2.1.0      2019-06-19 [3]
##  codetools      0.2-16     2018-12-24 [3]
##  colorblindr  * 0.1.0      2020-01-13 [2]
##  colorspace   * 1.4-2      2019-12-29 [2]
##  cowplot      * 1.0.0      2019-07-11 [2]
##  crayon         1.3.4      2017-09-16 [1]
##  data.table     1.12.8     2019-12-09 [2]
##  DBI            1.1.0      2019-12-15 [2]
##  dbplyr         2.0.0      2020-11-03 [1]
##  desc           1.2.0      2018-05-01 [2]
##  devtools       2.2.1      2019-09-24 [2]
##  digest         0.6.25     2020-02-23 [1]
##  dplyr        * 1.0.2      2020-08-18 [1]
##  ellipsis       0.3.1      2020-05-15 [1]
##  evaluate       0.14       2019-05-28 [2]
##  fansi          0.4.1      2020-01-08 [2]
##  farver         2.0.3      2020-01-16 [1]
##  fitdistrplus   1.0-14     2019-01-23 [2]
##  forcats      * 0.5.0      2020-03-01 [1]
##  formattable    0.2.0.1    2016-08-05 [1]
##  fs             1.5.0      2020-07-31 [1]
##  future         1.15.1     2019-11-25 [2]
##  future.apply   1.4.0      2020-01-07 [2]
##  gbRd           0.4-11     2012-10-01 [2]
##  gdata          2.18.0     2017-06-06 [2]
##  generics       0.0.2      2018-11-29 [2]
##  ggplot2      * 3.3.2      2020-06-19 [1]
##  ggrepel        0.8.1      2019-05-07 [2]
##  ggridges       0.5.2      2020-01-12 [2]
##  globals        0.12.5     2019-12-07 [2]
##  glue           1.3.2      2020-03-12 [1]
##  gplots         3.0.1.2    2020-01-11 [2]
##  gridExtra      2.3        2017-09-09 [2]
##  gtable         0.3.0      2019-03-25 [2]
##  gtools         3.8.1      2018-06-26 [2]
##  haven          2.3.1      2020-06-01 [1]
##  hms            0.5.3      2020-01-08 [1]
##  htmltools      0.4.0      2019-10-04 [2]
##  htmlwidgets    1.5.1      2019-10-08 [2]
##  httr           1.4.2      2020-07-20 [1]
##  ica            1.0-2      2018-05-24 [2]
##  igraph         1.2.5      2020-03-19 [1]
##  irlba          2.3.3      2019-02-05 [2]
##  jsonlite       1.7.1      2020-09-07 [1]
##  KernSmooth     2.23-16    2019-10-15 [3]
##  knitr          1.26       2019-11-12 [2]
##  labeling       0.3        2014-08-23 [2]
##  lattice        0.20-38    2018-11-04 [3]
##  lazyeval       0.2.2      2019-03-15 [2]
##  leiden         0.3.1      2019-07-23 [2]
##  lifecycle      0.2.0      2020-03-06 [1]
##  listenv        0.8.0      2019-12-05 [2]
##  lmtest         0.9-37     2019-04-30 [2]
##  lsei           1.2-0      2017-10-23 [2]
##  lubridate      1.7.9.2    2020-11-13 [1]
##  magrittr     * 2.0.1      2020-11-17 [1]
##  MASS           7.3-51.5   2019-12-20 [3]
##  Matrix         1.2-18     2019-11-27 [3]
##  memoise        1.1.0      2017-04-21 [2]
##  metap          1.2        2019-12-08 [2]
##  mnormt         1.5-5      2016-10-15 [2]
##  modelr         0.1.8      2020-05-19 [1]
##  multcomp       1.4-12     2020-01-10 [2]
##  multtest       2.42.0     2019-10-29 [2]
##  munsell        0.5.0      2018-06-12 [2]
##  mutoss         0.1-12     2017-12-04 [2]
##  mvtnorm        1.0-12     2020-01-09 [2]
##  nlme           3.1-143    2019-12-10 [3]
##  npsurv         0.4-0      2017-10-14 [2]
##  numDeriv       2016.8-1.1 2019-06-06 [2]
##  pbapply        1.4-2      2019-08-31 [2]
##  pillar         1.4.6      2020-07-10 [1]
##  pkgbuild       1.0.6      2019-10-09 [2]
##  pkgconfig      2.0.3      2019-09-22 [1]
##  pkgload        1.0.2      2018-10-29 [2]
##  plotly         4.9.1      2019-11-07 [2]
##  plotrix        3.7-7      2019-12-05 [2]
##  plyr           1.8.5      2019-12-10 [2]
##  png            0.1-7      2013-12-03 [2]
##  prettyunits    1.1.1      2020-01-24 [1]
##  processx       3.4.2      2020-02-09 [1]
##  progeny      * 1.11.3     2020-10-22 [1]
##  ps             1.3.2      2020-02-13 [1]
##  purrr        * 0.3.4      2020-04-17 [1]
##  R.methodsS3    1.7.1      2016-02-16 [2]
##  R.oo           1.23.0     2019-11-03 [2]
##  R.utils        2.9.2      2019-12-08 [2]
##  R6             2.4.1      2019-11-12 [1]
##  RANN           2.6.1      2019-01-08 [2]
##  rappdirs       0.3.1      2016-03-28 [2]
##  RColorBrewer   1.1-2      2014-12-07 [2]
##  Rcpp           1.0.4      2020-03-17 [1]
##  RcppAnnoy      0.0.16     2020-03-08 [1]
##  RcppParallel   4.4.4      2019-09-27 [2]
##  Rdpack         0.11-1     2019-12-14 [2]
##  readr        * 1.4.0      2020-10-05 [1]
##  readxl       * 1.3.1      2019-03-13 [2]
##  rematch        1.0.1      2016-04-21 [2]
##  remotes        2.1.0      2019-06-24 [2]
##  reprex         0.3.0      2019-05-16 [2]
##  reshape2       1.4.3      2017-12-11 [2]
##  reticulate     1.14       2019-12-17 [2]
##  rlang          0.4.8      2020-10-08 [1]
##  rmarkdown      2.0        2019-12-12 [2]
##  ROCR           1.0-7      2015-03-26 [2]
##  rprojroot      1.3-2      2018-01-03 [2]
##  rstudioapi     0.11       2020-02-07 [1]
##  rsvd           1.0.3      2020-02-17 [1]
##  Rtsne          0.15       2018-11-10 [2]
##  rvest          0.3.6      2020-07-25 [1]
##  sandwich       2.5-1      2019-04-06 [2]
##  scales         1.1.0      2019-11-18 [2]
##  sctransform    0.2.1      2019-12-17 [2]
##  SDMTools       1.1-221.2  2019-11-30 [2]
##  sessioninfo    1.1.1      2018-11-05 [2]
##  Seurat       * 3.1.2      2019-12-12 [2]
##  sn             1.5-4      2019-05-14 [2]
##  stringi        1.5.3      2020-09-09 [1]
##  stringr      * 1.4.0      2019-02-10 [1]
##  survival       3.1-8      2019-12-03 [3]
##  testthat       2.3.2      2020-03-02 [1]
##  TFisher        0.2.0      2018-03-21 [2]
##  TH.data        1.0-10     2019-01-21 [2]
##  tibble       * 3.0.4      2020-10-12 [1]
##  tidyr        * 1.1.2      2020-08-27 [1]
##  tidyselect     1.1.0      2020-05-11 [1]
##  tidyverse    * 1.3.0      2019-11-21 [2]
##  tsne           0.1-3      2016-07-15 [2]
##  usethis        1.5.1      2019-07-04 [2]
##  uwot           0.1.5      2019-12-04 [2]
##  vctrs          0.3.5      2020-11-17 [1]
##  viridis      * 0.5.1      2018-03-29 [2]
##  viridisLite  * 0.3.0      2018-02-01 [2]
##  withr          2.3.0      2020-09-22 [1]
##  xfun           0.12       2020-01-13 [2]
##  xml2           1.3.2      2020-04-23 [1]
##  yaml           2.2.1      2020-02-01 [1]
##  zoo            1.8-7      2020-01-10 [2]
##  source                                 
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  Bioconductor                           
##  Bioconductor                           
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  Github (clauswilke/colorblindr@1ac3d4d)
##  R-Forge (R 3.6.2)                      
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  Bioconductor                           
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  Github (saezlab/progeny@94bea1c)       
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.3)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
##  CRAN (R 3.6.2)                         
## 
## [1] /home/uhlitzf/R/lib
## [2] /usr/local/lib/R/site-library
## [3] /usr/local/lib/R/library

MSK SPECTRUM freeze major subset deep dives

Florian Uhlitz

2020-12-05

1 Cluster markers

1.1 Major Ovarian.cancer.cell markers for cell assign

1.2 Subtype currated markers

1.3 Subtype cluster markers

2 Clusters

2.1 sizes

2.2 UMAP

3 Filtering out doublet and mito high clusters

3.1 UMAP

3.2 add module scores and pathway scores

3.3 marker heatmap

3.4 composition

3.4.1 per site

3.4.2 per sample

3.4.3 site specific cluster enrichment

4 session info